Pulsar Astrophysics the Next Fifty Years Proceedings IAU Symposium No. 337, 2017 P. Weltevrede, B.B.P. Perera, L.L. Preston c International Astronomical Union 2018 & S. Sanidas, eds. doi:10.1017/S1743921317009668 The first binary pulsars and what they told us about binary evolution Dipankar Bhattacharya1 1 Inter-University Centre for Astronomy and Astrophysics Post Box 4, Ganeshkhind, Pune 411007, India email: [email protected] Abstract. The first few binary pulsars revealed the richness of evolution possible in binary systems containing neutron stars. Products of different evolutionary routes, in high and low mass binaries, as well as examples of evolution affected by the pulsar wind were among the first ten objects discovered. This article presents a historical review of the impact of binary pulsars on the early development of ideas regarding the evolution of neutron stars in binary systems. 1. Introduction Neutron stars in binary systems were first recognised as being members of bright, accreting X-ray binary sources [Zeldovich & Guseynov (1965)]. Several years of in- tense work on the evolution of such objects had followed when the first binary pulsar, PSR B1913+16, was discovered [Taylor & Hulse (1974); Hulse & Taylor (1975)]. This remarkable double neutron star system was immediately recognised as the product of X- ray binary evolution, and its properties revealed new aspects of the late stages of binary evolution. Lessons learnt from the first few binary pulsars discovered in the next decade and a half helped develop the current picture of the evolution of neutron stars in binary systems. In this article I will discuss this development, based on the first ten pulsars of this class listed in Table 1. The list includes nine pulsars with binary companions and the first millisecond pulsar PSR B1937+21, which though currently isolated, bears all signs of having been processed in a binary. Table 1. The first ten binary and millisecond pulsars, in order of their discovery Pulsar Pspin (s) log B surf (G) Porb (d) eccentricity M comp (M ) ref. B1913+16 0.059 10.358 0.323 0.617 1.44 1 B0655+64 0.196 10.068 1.029 7.5 × 10−6 ∼0.8 2 B0820+02 0.865 11.483 1232.404 0.012 ∼0.5 3 B1937+21 0.00156 8.612 – – – 4 B1953+29 0.00613 8.635 117.349 3.3 × 10−4 ∼0.2 5 B1831-00 0.521 10.874 1.811 0.0045 ∼0.07 6 B2303+46 1.066 11.896 12.339 0.658 ∼1.4 6 B1820-11 0.280 11.799 357.762 0.795 ∼0.8 7 B1855+09 0.00536 8.496 12.327 2.2 × 10−5 0.27 8 B1957+20 0.0161 8.223 0.382 ∼0.0 ∼0.025 9 Notes: The surface magnetic field B surf above is the dipole component estimated from spin period and spin-down rate. Companion masses M comp are derived from relativistic orbital parameters for B1913+16 and B1855+09, while in other cases they are estimates based on the mass function, assuming a pulsar mass of 1.4 M and an inclination of 60◦. For B0820+02 the mass is constrained by optical observations. Discovery references: 1. Hulse & Taylor (1975), 2. Damashek, Taylor & Hulse (1978), 3. Manchester et al.(1978), 4. Backer et al.(1982), 5. Boriakoff et al.(1983), 6. Dewey et al.(1985), 7. Clifton & Lyne (1986), 8. Segelstein et al.(1986) 9. Fruchter et al.(1988) 37 Downloaded from https://www.cambridge.org/core. IP address: 170.106.40.219, on 25 Sep 2021 at 14:05:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1743921317009668 38 D. Bhattacharya From the first binary pulsar system B1913+16 alone, several important conclusions could be drawn, namely: • A combination of low magnetic field strength and fast spin suggested spin-up of the neutron star by accretion [Smarr & Blandford (1976)]. • The small orbital separation suggested that the binary system went through a common-envelope evolution [Smarr & Blandford (1976)]. • A bound orbit of two neutron stars implied that the binary system remained bound after two supernovae. This suggested that neutron stars may receive substantial kicks at birth [Flannery & van den Heuvel (1975)]. • Orbital shrinkage due to gravitational radiation was measured, highlighting the im- portance of gravitational radiation in the evolution of compact star binary systems [Weis- berg & Taylor (1981)]. Lessons derived from subsequent discoveries listed in Table 1 include: • The existence of a new class of pulsars with low-mass binary companions • Diverse evolutionary pathways of low-mass donor systems • Reinforcement of the spin-up scenario • Irradiation and evaporation of the companion of a pulsar • Evolution of the magnetic fields of neutron stars In the following sections I will briefly describe some of the issues mentioned above. Much of the discussion presented here may be found elaborated in the review article Bhattacharya & van den Heuvel (1991). 2. Spin-up by accretion Even before the discovery of the first binary pulsar, Davidson & Ostriker (1973) had discussed in detail the idea of neutron star spin being affected by accretion of matter in a binary system. The inflowing matter forms an accretion disc, which is truncated at an inner radius, called the“Alfv´en radius”, determined by the magnetic field strength of the neutron star. Below this radius the accreting plasma is strongly coupled to the magnetic field and co-rotates with the neutron star. A spin equilibrium is achieved if the keplerian frequency at the Alfv´en radius equals the neutron star’s spin rate. If the neutron star is spinning slower than that, then the accreting matter would tend to spin it up, and vice versa. The lower the magnetic field, the closer to the neutron star the Alfv´en radius will be, and the shorter will be the corresponding equilibrium spin period. PSR B1913+16 distinguished itself from the rest of the known pulsar population at that time by its relatively low magnetic field strength (about two orders of magnitude below the average pulsar field), and a fairly fast spin period of 59 ms. Smarr & Blandford (1976) concluded that this combination implies that the neutron star must have experienced accretion in the binary system and has thereby been spun-up. The lower than average field strength allowed the spin-up process to achieve the observed short period. For a given magnetic field strength B, the equilibrium period achieved in the spin-up process depends on the mass accretion rate M˙ – the higher the mass inflow rate, the smaller is the Alfv´en radius. For a dipole magnetic geometry the equilibrium spin period 2 3/7 has the dependence Peq ∝ (B /M˙ ) . It was pointed out by Srinivasan & van den Heuvel (1982) that this implies a lower limit to the period to which a neutron star of a given field strength could be spun up, corresponding to the maximum accretion rate, namely −8 the Eddington rate M˙ Edd ∼ 10 M/yr. This relation between Peq and B, keeping M˙ fixed at M˙ Edd was named the “Critical spin-up line”, and is referred to as simply the “Spin-up line” in current literature. Downloaded from https://www.cambridge.org/core. IP address: 170.106.40.219, on 25 Sep 2021 at 14:05:44, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1743921317009668 First Binary Pulsars 39 In a paper presented at the 1981 Asia Pacific Regional Meeting in Bandung, Indonesia, Radhakrishnan & Srinivasan (1984) argued that a whole population of spun-up pulsars must exist, and in a plot of pulsar magnetic field vs spin period, such objects would be found in the region bordered by the spin-up line on the left and the “Death Line” on the right. They called this population “Recycled Pulsars”, a term still in current use. Subsequent discoveries of binary pulsars have amply borne out this prediction. The discovery of the millisecond pulsar PSR B1937+21 [Backer et al.(1982)] put a spotlight on the recycling scenario. Here was a pulsar with a very short period and very low field, for which recycling would be a perfect explanation [Radhakrishnan & Srinivasan (1982); Alpar et al.(1982)], but the lack of a companion was a problem. However, soon afterwards similar pulsars with binary companions were found [Boriakoff et al.(1983); Segelstein et al.(1986)], confirming the spin-up origin for millisecond pulsars. 3. Orbital Evolution - mass transfer, spiral-in, supernova explosion With two neutron stars in the system, it was recognised that PSR B1913+16 must have descended from a high-mass X-ray binary system, in which the massive progenitor of the second born neutron star transferred mass onto the recycled pulsar. The orbital separation of the final system (periastron distance ∼ 1 R) is however too small to fit the mass donating progenitor. Rapid shrinkage of orbit must therefore have taken place in the final stages of mass transfer. This implied highly non-conservative evolution of the binary in the late stages, via either a spiral-in process in a common envelope [Smarr & Blandford (1976)] or a major outflow driven by super-Eddington mass transfer [Flannery & van den Heuvel (1975)]. This sparked detailed radiation-hydrodynamic treatment of common envelope evolution, beginning with Bodenheimer & Taam (1984) and continuing till today. In some cases common envelope evolution may lead to a complete merger of the binary components, leaving an isolated recycled pulsar [Bhattacharya & van den Heuvel (1991); van den Heuvel & Bonsema (1984)], but clear observational signature of this is difficult to find.